prepare scientific manuscripts

Researchers face challenges in academic writing—including verbose LaTeX coding, poor version control, low collaborative efficiency, and irreproducible results. To address these, this paper proposes a GitHub-based automated manuscript generation framework. It adopts Markdown as the source format and integrates automated LaTeX compilation, programmatic figure generation, and continuous integration (CI) pipelines to enable end-to-end, traceable, and reproducible scientific writing—from raw data to publication-ready PDF. Crucially, the framework redefines manuscripts as “executable outputs,” unifying version control, computational provenance tracking, and dynamic content updating. Evaluated in computational biology and microscopic image analysis workflows, the framework significantly improves collaborative productivity and adherence to open science principles, while supporting high-quality, fully reproducible scholarly publishing.

Scientific posters, as vital vehicles for scholarly communication, have long suffered from low sharing rates, the absence of persistent identifiers, incomplete metadata, and a lack of established citation practices, all of which hinder their discoverability and reuse. This study presents the first large-scale empirical analysis of 86 global poster-sharing platforms, integrating platform surveys, metadata assessments, and usage metrics—including views, downloads, and citations—with a focus on major repositories such as Zenodo and Figshare. As of 2024, only approximately 150,000 posters are publicly shared, with most platforms failing to adhere to FAIR principles, exhibiting critical gaps in conference-related metadata and extremely low citation rates. The findings reveal a disconnect between platform capabilities and user behaviors, prompting the proposal of community guidelines to standardize poster sharing and reuse.

The use of visual embellishments in scientific papers is increasingly common, yet their impact on information communication and reader experience remains unclear. This study systematically reviews 374 visualization papers published between 2019 and 2024 in IEEE VIS, ACM CHI, and EuroVis that incorporate decorative elements. It introduces a novel three-dimensional analytical framework—Purpose, Design, and Placement—to characterize how such embellishments are employed. Through a combination of systematic literature review and qualitative content analysis, the work identifies prevalent usage patterns and potential functional roles of visual decorations in scholarly communication. By offering a structured perspective on the role of embellishment in scientific publishing, this research provides empirical grounding for evolving norms in scientific writing and visualization design practices.

Scientific manuscripts often exhibit inconsistent and non-portable visual representations of technical insights due to ad hoc formatting. To address this, we propose a modular, semantic LaTeX framework that abstracts visual elements—such as color schemes, highlight boxes, classification trees, and author metadata—into lightweight, namespace-scoped packages (e.g., `ktcolor`, `ktbox`, `ktlrtree`, `ktorcid`). These packages support automatic numbering, full-width highlighting, multi-column layouts, and embedded content. By decoupling presentation style from document structure, the framework enables semantics-driven visualization while maintaining compatibility with major document classes—including IEEEtran, ACM’s `acmart`, and Beamer. Compared to manual typesetting, our approach significantly improves consistency, maintainability, and information conveyance efficiency in academic writing. It establishes a reusable, extensible infrastructure for structured knowledge representation in technical documents.

To address the lack of structured guidance for academic poster layout design amid the explosive growth of scholarly publications, this paper introduces SciPostGen—the first large-scale dataset of paper–poster layout pairs—and systematically uncovers statistical correlations between paper structural features (e.g., section count, figure/table quantity) and spatial distributions of layout elements. Building upon this insight, we propose a retrieval-augmented generation (RAG) framework: it first retrieves semantically similar historical layouts as structural priors, then jointly conditions layout generation on both the input paper’s content and explicit structural constraints. Experiments demonstrate that our method accurately predicts layouts tailored to paper structure, producing high-fidelity, semantically coherent poster designs—both with and without explicit constraints—outperforming all baseline models significantly. This work establishes a novel paradigm and foundational infrastructure for automating academic visualization.

This study addresses the underexplored yet critical issue of academic misconduct orchestrated by paper mills within conference proceedings, which poses a severe threat to research integrity. For the first time, it systematically uncovers the large-scale infiltration of paper mills into IEEE international conferences. By collecting over 4,000 ghostwriting advertisements from social media and applying a combination of semi-automated text matching, manual verification, and multidimensional anomaly detection—including authorship collaboration patterns, institutional diversity, citation manipulation, and content analysis—the study identifies 1,720 problematic papers across 286 conferences, involving more than 6,500 authors and over 3,500 institutions from 55 countries, with a single conference exhibiting a peak prevalence of 23.51%. The findings reveal that paper mills operate in a highly organized, openly commercialized, and transnationally collaborative manner, providing empirical grounding for strengthening academic integrity governance.

Existing automated chart generation systems are limited to a single chart type and plain-text input, producing non-editable bitmap outputs that fail to meet the demands of scientific research for diverse inputs and high-quality, editable visualizations. To address this, this work proposes Crafter, a multi-agent framework that leverages a novel collaborative multi-agent mechanism to uniformly handle multiple chart types and cross-modal inputs without architectural modifications. Complementing Crafter, CraftEditor converts bitmap outputs into structured SVGs with high fidelity, enabling fine-grained local editing. Evaluated on PaperBanana-Bench and the newly introduced CraftBench, the proposed approach significantly outperforms existing methods, achieving state-of-the-art SVG generation quality across all baselines, with ablation studies confirming the effectiveness of each component.

本文介绍COCI框架，利用AI技术从征文启事中提取结构化元数据，解决灰色文献难以集成到学术知识图谱的问题。

This work addresses the challenge of automatically transforming static scientific papers into dynamic presentation formats—such as posters, slides, and videos—while preserving semantic consistency across modalities. We formalize this as the Unified Presentation Suite Generation task and propose a centralized content planning framework grounded in renderable HTML. To ensure coherence among multimodal outputs, we introduce a self-correcting verification-and-repair loop. Our contributions include OmniPreBench, a large-scale dataset; a vision-language model–based evaluation protocol; and a method for aligning multimodal content. Experimental results demonstrate that our approach significantly outperforms strong baselines in both factual accuracy and visual appeal, enabling the automatic generation of high-quality, semantically consistent scientific communication materials.

This work addresses the challenges of structural inconsistency, missing content, and cross-section incoherence commonly encountered in automatically generated scientific papers, particularly between narrative text, experimental evidence, and visual elements. To resolve these issues, the authors propose a multi-agent collaborative framework grounded in a persistent shared visual contract, comprising architect, writer, optimizer, renderer, and evaluator agents. These agents operate within a generate–evaluate–adapt loop that dynamically updates the contract to align textual structure with visual components throughout the document. The approach introduces, for the first time, a contract-driven mechanism to regulate multi-agent collaboration. Evaluated on the Jericho corpus, the method achieves an expert rating of 6.145, significantly outperforming DirectChat (3.963) and Fars (5.197), thereby demonstrating substantial improvements in both structural coherence and text–figure consistency.